|By Mark Little||
|October 21, 2002 12:00 AM EDT||
Use of atomic transactions is a well-known technique for guaranteeing consistency in the presence of failures. The ACID properties of atomic transactions (Atomicity, Consistency, Isolation, Durability) ensure that even in complex business applications consistency of state is preserved.
Transactions are best viewed as "short-lived" entities operating in a closely coupled environment, performing stable state changes to the system; they are less well suited for structuring "long-lived" application functions (e.g., running for hours, days, etc.) and running in a loosely coupled environment like the Web. Long-lived atomic transactions (as typically occur in business-to-business interactions) may reduce the concurrency in the system to an unacceptable level by holding on to resources for a long time; further, if such an atomic transaction rolls back, much valuable work already performed could be undone. As a result, there have been various extended transactions models where strict ACID properties can be relaxed in a controlled manner. Until recently, translating these models into the world of Web services had not been attempted. However, the OASIS Business Transaction Protocol, specified by a collaboration of several companies, has tried to address this issue. In this article, the second in a two-part series, we'll describe how the BTP has attempted to solve these problems.
Architecture of the Business Transaction Protocol
A very high-level view of the BTP can be described as follows: Web services do work within the scope of atoms, which are created by the initiator of a business transaction; multiple atoms are composed into a business transaction (e.g., arranging a holiday) by a cohesion composer such that different atoms may possess different outcomes, as directed by the business logic, e.g., cancel one insurance quote and confirm another. Businesses take part in atomic or cohesive transactions via participants, and both cohesions and atoms use coordination to ensure that participants see the desired outcome (see Figure 1). This may seem fairly straightforward at first, but as we shall see in the following sections, there's a lot more going on under the covers!
The XML Context
In order for a transaction to span a distributed number of services/tasks, certain information has to flow between the sites/domains involved in the application. This is commonly referred to as the context and typically contains the following information:
The context information is propagated to provide a flow of context information between distributed execution environments, for example using SOAP header information. This may occur transparently to the client and application services. The context is propagated as part of normal message interchange within an application (e.g., as an additional part of the SOAP header).
XML Message Sets and Carrier Bindings
In the Web services world, information is communicated in XML documents, but how those documents are exchanged may be a function of the environment, business relationship, etc. Therefore, although BTP mandates that its own information (context and protocol messages) must be carried in XML payloads, it doesn't specify how these payloads are transmitted; it doesn't mandate a specific carrier protocol.
Obviously, without a carrier protocol, BTP is of very limited use! The technical committee did define a binding to SOAP 1.1 over HTTP 1.1 as part of the BTP 1.0 specification, but the intention has always been that other specific carrier protocol bindings to the BTP XML schema would be provided on an as-needed basis. So if, for example, a group of companies sees merit in defining a binding using pigeons(!), they could so define it and submit it as an appendix on optional bindings to the BTP specification.
As with traditional transaction processing systems, the BTP message set is concerned with messages for driving the protocol and messages containing information for participating within the protocol. The former are typically of interest only to implementers of either BTP or participants, whereas the latter are of interest to service providers and their associated participants.
Typically a BTP message is propagated within the body of the SOAP envelope. For example, Listing 1 shows a typical begin message.
For application messages that also carry BTP content, the situation is different. In this situation the BTP messages are typically located within the header of the SOAP envelope, as can be seen in Listing 2, in which a BTP context is propagated with an application-specific method call.
The Web Service
Whenever a user contacts a Web service, e.g., a taxi booking service, whose work it wishes to be under the control of a transaction, components of the transaction system are responsible for flowing the context to that service. The service can then use this information to enlist a participant with the transaction. The service is responsible for ensuring that concurrent accesses by different applications are managed in a way that guarantees some internal consistency criteria for that service. Note that a Web service may also play the role of a participant.
The participant is the entity that does the real transaction work. The Web service (e.g., a theater booking system) contains some business logic for reserving a seat, inquiring about availability, etc., but it will need to be back-ended by something that maintains information in a durable manner. Typically this will be a database, but it could be a file system, NVRAM, etc.
Now, although the service may talk to the back-end database directly, it cannot commit or roll back any changes it (the service) makes, since these are ultimately under the control of the transaction that scoped the work. In order for the transaction to be able to exercise this control, it must have some contact with the back-end resource (the database in our example), and this is accomplished by the participant.
Each participant supports a two-phase termination protocol via the prepare, confirm, and cancel operations. What the participant does when asked to prepare is implementation dependent (e.g., reserve the theater ticket); it then returns an indication of whether or not it succeeded. However, unlike in an atomic transaction, the participant does not have to guarantee that it can remain in this prepared state; it may indicate that it can only do so for a specified period of time, and also indicate what action it will take (confirm or undo) if it has not been told how to finish before this period elapses. In addition, no indication of how the prepare is implemented is implied in the protocol, so resource reservation (locking), as happens in an ACID transaction system, need not occur.
Associated with every transaction type (atom or cohesion) is a coordinator, which is responsible for governing the outcome of the transaction. The coordinator may be implemented as a separate service or may be colocated with the user for improved performance. It communicates with enlisted participants to inform them of the desired termination requirements, i.e., whether they should accept (confirm) or reject (cancel) the work done within the scope of the given transaction. For example, whether to purchase the (provisionally reserved) flight tickets for the user or to release them. This communication will be an implementation-specific protocol (e.g., two- or three-phase completion).
A transaction manager factory is typically responsible for managing coordinators for many transactions. The initiator of the transaction (e.g., the client) communicates with a transaction manager and asks it to start a new transaction and associate a coordinator with the transaction. Once created, the context can be propagated to Web services in order for them to associate their work with the transaction.
The atom coordinator is typically used to scope work performed on Web services. The cohesion composer is the business logic for gluing together the flow of the application into one or more atoms. Although Web services do work within the scope of a specific atom, it is the composer that ultimately determines which atoms to confirm, and which to undo; as participants are to atoms, so atoms are to cohesion composers (cohesions). The composer may prepare and cancel atoms at arbitrary points during the lifetime of the business transaction, e.g., preparing the flight reservation early in the transaction, and preparing the insurance quote much later after cancelling a prior quote. The main difference between an atom and a cohesion is that whereas all participants enrolled with an atom will either confirm or cancel, the participants enrolled with a cohesion (multiple atoms) may have different outcomes. However, once the composer has arrived at its confirm set (the participants that will confirm), it essentially collapses down to become an atom and guarantees an all-or-nothing effect, i.e., all atoms in the confirm set will either confirm or cancel, with no intermediate effects.
Superiors and Inferiors
Although for simplicity we've talked about services, coordinators, and participants, within BTP all end points are either Superiors or Inferiors or both. An actor within the coordinating entity's system plays the role of Superior (e.g., the atom coordinator) and an actor within the service plays the role of an Inferior (e.g., the participant). Each Inferior has only one Superior. However, a single Superior may have multiple Inferiors within single or multiple parties. A tree of such relationships may be wide, deep, or both, as shown in Figure 2.
An Inferior is typically associated with some set of application activities. Usually this will be a result of some operation invocations (on a "service application element") from elsewhere (an "initiating application element"). The Inferior is responsible for reporting to the Superior that it is "prepared" for the outcome whether or not the associated operations' provisional effect can be confirmed or cancelled.
A Superior receives reports from its Inferiors as to whether they are prepared to give an outcome. It gathers these reports in order to determine which Inferiors should be canceled and which confirmed. The Superior does this either by itself or with the cooperation of the application element responsible for its creation and control, depending upon whether the transaction is an atom or a cohesion, as we shall see later.
The initiator of the atom communicates with an atom/cohesion manager (factory) and asks it to start a new atom. Once created, information about the atom or cohesion (the context) can be propagated to Web services in order for them to associate their work with it. Although work is typically conducted within the scope of an atom, it is entirely possible for services to register participants directly with cohesions.
The terminator of the atom or cohesion will typically be the same entity as the initiator, but need not be. For example a long-running stock purchase transaction may be started by the company that requires the stock, and finished by the company that delivers it. Although an atom can be instructed to confirm all participants immediately, it is more typically instructed to prepare them first, and later (hours, days, etc.) to either confirm or cancel them.
BTP gives builders of transactional Web services the ability to concentrate on the functional aspects of their services (e.g., what it means to book an airline ticket), and to guarantee consensus through the participant interface. Since the participant interface is transparent to its implementation, a provider may use any implementation appropriate to the Web service it acts on behalf of.
Through the cohesion composer, BTP gives the business logic the flexibility to structure interactions with services into multiple (dynamic) consensus groups. The important distinction between BTP and atomic transactions is that multiple such groups exist in BTP, compared to one in atomic transactions, and the cohesion has the capability to drive the two-phase termination protocol explicitly. The fact that atoms may be prepared at any point in the normal flow of business, and later confirmed or undone, gives greater flexibility to the application.
We have described how BTP can be used to conduct typical business-to-business interactions in a reliable manner. In order to do this, many protocol-specific messages need to be exchanged between actors, and this will have an adverse effect on the time taken to complete a business transaction. This is a necessary side effect of achieving reliability and consensus and is not specific to BTP.
Since BTP is intended for long-running transactions, it may be assumed that performance hasn't been a prime factor in its development. However, this is not the case and, in fact, BTP contains a number of optimizations.
Typically a participant is enlisted with a BTP transaction when a service invocation occurs (e.g., "book flight"). When the service request completes, the response is sent back to the initiator of the request. As described earlier, during transaction termination the coordinator will interact with the participant to ensure completion.
In some circumstances it may be possible to compound many of the above messages into a "one-shot" message. For example, the service invocation may cause a state change to occur that means the participant can prepare immediately after the invocation completes. Rather than having to wait for an explicit coordinator message, BTP allows the enroll request and statement of preparation to be compounded within the service response. The receiver is then responsible for ensuring that this additional information is forwarded to the responsible actors.
Resignation by Participant
In a two-phase commit protocol, in addition to indicating success or failure during the preparation phase, a participant can also return a "read-only" response; this indicates that it doesn't control any work that has been modified during the course of the transaction and therefore doesn't need to be informed of the transaction outcome. In some situations this allows the two-phase protocol to complete quickly, since a second round of messages isn't required.
The equivalent of this in BTP is for a participant to resign from the transaction it was enrolled in. Resignation can occur at any time up to the point at which the participant has prepared. Resignation is used by the participant to indicate that it no longer has an interest in the outcome of the transaction.
In some situations, rather than waiting for an instruction from the coordinator to prepare, a participant may be able to spontaneously prepare. For example, a service invocation occurs, moving the service into an idempotent state such that further invocations have no effect on it; in this case, an associated participant may prepare the service immediately, rather than wait for the instruction to do so. In BTP, a participant is allowed to attempt to prepare at any point and inform the coordinator of the result.
Autonomous Decision by Participant
In a traditional two-phase protocol a participant enrolls with a transaction and waits for the termination protocol before it either confirms or cancels. To achieve consensus, it is necessarily a blocking protocol, which means that if a coordinator fails before delivering the final phase messages, prepared participants must remain blocked, holding on to (possibly valuable) resources. Modern transaction-processing systems have augmented the two-phase commit with heuristics, which allow such participants to make unilateral decisions about whether they will commit or roll back. Obviously if a participant makes a choice that turns out to be different from that of other participants, nonatomic behavior occurs.
BTP has its equivalent of heuristics, allowing participants to make unilateral decisions as well. However, unlike other transaction implementations, the protocol allows a participant to give the coordinator prior knowledge of what the decision will be and when it will occur. A participant may prepare and present the coordinator with some caveats as to how long it will remain in this state and into what state it will then migrate (e.g., "will remain prepared for 10 days and then will cancel the flight reservation"). This information may then be used by the coordinator to optimize message exchange.
BTP and the Web Services Stack
So where exactly does BTP fit into the evolving Web services architecture? As shown in Figure 3, it is primarily intended as a low-level protocol, hidden from users in much the same way traditional transaction systems are. Typically, a user would see just a demarcation API (e.g., how to start and end an atom); the BTP specification does not define any such API because it is language independent. One possible API that readers should be aware of is that being developed in JSR 156 - Java API for XML Transactions.
So How Would I Use This BTP Thing?
Consider the flight booking example presented earlier. How could we use BTP in order to coordinate this application in a reliable manner? The problem is that we wish to obtain the cheapest insurance quote as we go along, without losing prior quotes until we know that they are no longer the cheapest; at that point we will be able to release those quotes while maintaining others. In a traditional transaction system, all of the work performed within a transaction must either be accepted (committed) or declined (rolled back); the required loosening of atomicity is not supported.
In BTP, however, we can use atoms and cohesions. A cohesion is first created to manage the overall business interactions. The business logic (application, client, etc.) creates an atom (i.e., ReserveAtom) and enrolls it with the cohesion, as shown in Figure 4.
Once the client has obtained the context from the factory, it can invoke the airline and taxi reservation services within the scope of the atom, such that their work is then ultimately controlled by its outcome. When a suitable flight and taxi can be obtained, ReserveAtom is prepared to reserve the bookings for some service-specific time.
Then two new atoms (AtomQuote1 and AtomQuote2) are created and enrolled with the cohesion, before being used to obtain two different quotes from the respective insurance services.
When the quote from the first insurance site is obtained it is obviously not known whether it is the best quote, so the business logic can prepare AtomQuote1 to maintain the quote, while it then communicates with the second insurance site. If that site does not offer a better quote, the application can cancel AtomQuote2 and it now has its final confirmation set of atoms (ReserveAtom and AtomQuote1), which it can confirm (see Figure 5).
ACID transactions have proven invaluable over the years in the construction of enterprise applications. However, they are only really suited to short-duration activities executing on closely coupled applications and environments. When used in a loosely coupled environment, they prove too inflexible and restricting for many applications. The OASIS Business Transactions Protocol has been developed to solve this problem while at the same time maintaining those aspects of the atomic transaction model that have proven useful. At the time of this writing, there is only a single BTP implementation available, from Hewlett-Packard. However, several companies have stated that they are working on their own implementations.
SYS-CON Events announced today that AppNeta, the leader in performance insight for business-critical web applications, will exhibit and present at SYS-CON's @DevOpsSummit at Cloud Expo New York, which will take place on June 7-9, 2016, at the Javits Center in New York City, NY. AppNeta is the only application performance monitoring (APM) company to provide solutions for all applications – applications you develop internally, business-critical SaaS applications you use and the networks that deli...
Feb. 14, 2016 11:45 AM EST Reads: 454
For Valentine's Day, here's a lighthearted look at the "relationship" between two complementary technologies: service virtualization and cloud dev/test labs. Hey, I know it's been a while since we started being "a thing." When we met, everyone said you were just mocking, and that I wasn't real enough to make a living, with my head in the clouds. Yet, here we are, a few years later.
Feb. 14, 2016 11:45 AM EST
SYS-CON Events announced today that Column Technologies will exhibit at SYS-CON's @DevOpsSummit at Cloud Expo, which will take place on June 7-9, 2016, at the Javits Center in New York City, NY. Established in 1998, Column Technologies is a global technology solutions provider with over 400 employees, headquartered in the United States with offices in Canada, India, and the United Kingdom. Column Technologies provides “Best of Breed” technology solutions that automate the key DevOps principal...
Feb. 14, 2016 11:45 AM EST
More and more companies are looking to microservices as an architectural pattern for breaking apart applications into more manageable pieces so that agile teams can deliver new features quicker and more effectively. What this pattern has done more than anything to date is spark organizational transformations, setting the foundation for future application development. In practice, however, there are a number of considerations to make that go beyond simply “build, ship, and run,” which changes ho...
Feb. 14, 2016 11:45 AM EST Reads: 289
Adding public cloud resources to an existing application can be a daunting process. The tools that you currently use to manage the software and hardware outside the cloud aren’t always the best tools to efficiently grow into the cloud. All of the major configuration management tools have cloud orchestration plugins that can be leveraged, but there are also cloud-native tools that can dramatically improve the efficiency of managing your application lifecycle. In his session at 18th Cloud Expo, ...
Feb. 14, 2016 09:30 AM EST Reads: 105
Microservices are a type of software architecture where large applications are made up of small, self-contained units working together through APIs that are not dependent on a specific language. Each service has a limited scope, concentrates on a specific task and is highly independent. This setup allows IT managers and developers to build systems in a modular way. In his book, “Building Microservices,” Sam Newman said microservices are small, focused components built to do a single thing very w...
Feb. 14, 2016 09:30 AM EST Reads: 116
When building large, cloud-based applications that operate at a high scale, it’s important to maintain a high availability and resilience to failures. In order to do that, you must be tolerant of failures, even in light of failures in other areas of your application. “Fly two mistakes high” is an old adage in the radio control airplane hobby. It means, fly high enough so that if you make a mistake, you can continue flying with room to still make mistakes. In his session at 18th Cloud Expo, Lee...
Feb. 14, 2016 09:15 AM EST
WebSocket is effectively a persistent and fat pipe that is compatible with a standard web infrastructure; a "TCP for the Web." If you think of WebSocket in this light, there are other more hugely interesting applications of WebSocket than just simply sending data to a browser. In his session at 18th Cloud Expo, Frank Greco, Director of Technology for Kaazing Corporation, will compare other modern web connectivity methods such as HTTP/2, HTTP Streaming, Server-Sent Events and new W3C event APIs ...
Feb. 14, 2016 08:30 AM EST
Join us at Cloud Expo | @ThingsExpo 2016 – June 7-9 at the Javits Center in New York City and November 1-3 at the Santa Clara Convention Center in Santa Clara, CA – and deliver your unique message in a way that is striking and unforgettable by taking advantage of SYS-CON's unmatched high-impact, result-driven event / media packages.
Feb. 14, 2016 08:30 AM EST Reads: 119
At first adopted by enterprises to consolidate physical servers, virtualization is now widely used in cloud computing to offer elasticity and scalability. On the other hand, Docker has developed a new way to handle Linux containers, inspired by version control software such as Git, which allows you to keep all development versions. In his session at 17th Cloud Expo, Dominique Rodrigues, the co-founder and CTO of Nanocloud Software, discussed how in order to also handle QEMU / KVM virtual machin...
Feb. 14, 2016 08:15 AM EST Reads: 164
How is your DevOps transformation coming along? How do you measure Agility? Reliability? Efficiency? Quality? Success?! How do you optimize your processes? This morning on #c9d9 we talked about some of the metrics that matter for the different stakeholders throughout the software delivery pipeline. Our panelists shared their best practices.
Feb. 14, 2016 08:00 AM EST Reads: 133
Microservices are all the rage right now — and the industry is still learning, experimenting, and developing patterns, for successfully designing, deploying and managing Microservices in the real world. Are you considering jumping on the Microservices-wagon? Do Microservices make sense for your particular use case? What are some of the “gotchas” you should be aware of? This morning on #c9d9 we had experts from popular chat app Kik, SMB SaaS platform Yodle and hosted CI solution Semaphore sha...
Feb. 14, 2016 08:00 AM EST
SYS-CON Events announced today that FalconStor Software® Inc., a 15-year innovator of software-defined storage solutions, will exhibit at SYS-CON's 18th International Cloud Expo®, which will take place on June 7-9, 2016, at the Javits Center in New York City, NY. FalconStor Software®, Inc. (NASDAQ: FALC) is a leading software-defined storage company offering a converged, hardware-agnostic, software-defined storage and data services platform. Its flagship solution FreeStor®, utilizes a horizonta...
Feb. 14, 2016 07:30 AM EST
Sensors and effectors of IoT are solving problems in new ways, but small businesses have been slow to join the quantified world. They’ll need information from IoT using applications as varied as the businesses themselves. In his session at @ThingsExpo, Roger Meike, Distinguished Engineer, Director of Technology Innovation at Intuit, showed how IoT manufacturers can use open standards, public APIs and custom apps to enable the Quantified Small Business. He used a Raspberry Pi to connect sensors...
Feb. 14, 2016 04:30 AM EST Reads: 408
Father business cycles and digital consumers are forcing enterprises to respond faster to customer needs and competitive demands. Successful integration of DevOps and Agile development will be key for business success in today’s digital economy. In his session at DevOps Summit, Pradeep Prabhu, Co-Founder & CEO of Cloudmunch, covered the critical practices that enterprises should consider to seamlessly integrate Agile and DevOps processes, barriers to implementing this in the enterprise, and pr...
Feb. 14, 2016 04:00 AM EST Reads: 495
The cloud promises new levels of agility and cost-savings for Big Data, data warehousing and analytics. But it’s challenging to understand all the options – from IaaS and PaaS to newer services like HaaS (Hadoop as a Service) and BDaaS (Big Data as a Service). In her session at @BigDataExpo at @ThingsExpo, Hannah Smalltree, a director at Cazena, will provide an educational overview of emerging “as-a-service” options for Big Data in the cloud. This is critical background for IT and data profes...
Feb. 14, 2016 04:00 AM EST Reads: 264
If we look at slow, traditional IT and jump to the conclusion that just because we found its issues intractable before, that necessarily means we will again, then it’s time for a rethink. As a matter of fact, the world of IT has changed over the last ten years or so. We’ve been experiencing unprecedented innovation across the board – innovation in technology as well as in how people organize and accomplish tasks. Let’s take a look at three differences between today’s modern, digital context...
Feb. 14, 2016 03:45 AM EST Reads: 262
The principles behind DevOps are not new - for decades people have been automating system administration and decreasing the time to deploy apps and perform other management tasks. However, only recently did we see the tools and the will necessary to share the benefits and power of automation with a wider circle of people. In his session at DevOps Summit, Bernard Sanders, Chief Technology Officer at CloudBolt Software, explored the latest tools including Puppet, Chef, Docker, and CMPs needed to...
Feb. 14, 2016 03:30 AM EST Reads: 399
The (re?)emergence of Microservices was especially prominent in this week’s news. What are they good for? do they make sense for your application? should you take the plunge? and what do Microservices mean for your DevOps and Continuous Delivery efforts? Continue reading for more on Microservices, containers, DevOps culture, and more top news from the past week. As always, stay tuned to all the news coming from@ElectricCloud on DevOps and Continuous Delivery throughout the week and retweet/favo...
Feb. 14, 2016 03:00 AM EST Reads: 257
In a previous article, I demonstrated how to effectively and efficiently install the Dynatrace Application Monitoring solution using Ansible. In this post, I am going to explain how to achieve the same results using Chef with our official dynatrace cookbook available on GitHub and on the Chef Supermarket. In the following hands-on tutorial, we’ll also apply what we see as good practice on working with and extending our deployment automation blueprints to suit your needs.
Feb. 14, 2016 02:00 AM EST Reads: 107